Objectives To assess the baseline distribution and prognostic capacity of MRI lesions in the SIJ of patients diagnosed with axSpA after 2 years follow up in the ECHOSPA cohort.

Methods Consecutive outpatients with age <50 years and symptoms>3 months suggestive of SpA (inflammatory back pain, peripheral arthritis or inflammatory arthralgia, enthesitis or dactylitis, uveitis with B27 positivity, a family history of SpA) were enrolled in the prospective French ECHOSPA cohort study. The diagnosis of SpA was ascertained by an expert committee, blind to MRI evaluation, after at least 2 years of follow-up. MRI scans from 223 cases were available for evaluation by 2 readers and an adjudicator who assessed MRI lesions in the SIJ according to updated consensus definitions from the ASAS-MRI group. These were recorded in an ASAS consensus-derived eCRF that comprises global assessment (active and/or structural lesion typical of axSpA present/absent) and detailed scoring of individual lesions (SPARCC SIJ inflammation, SPARCC SIJ structural). Clinical, lab, and imaging variables associated with the diagnosis of axSpA at 2 years were first identified by univariate regression. A base model of all clinical/lab variables associated with axSpA was included as a group in multivariate logistic regression models that tested the independent association with MRI lesions.

Results Mean age of the 223 cases was 39.6 (10.5) years, mean symptom duration was 2.5 (4.1) years, 49.5% were HLA-B27 +and 63.7% were female. Primary inclusion criterion was IBP in 53%, IA in 27%, ED in 9%, B27 +U in 8% and Fam in 4%. At 2 years follow up, 165 (74%) were deemed to have axSpA. In group comparisons (table 1) and univariate regression, both active and structural MRI lesions were associated with diagnosis of axSpA at 2 years (p=0.03, p=0.01, respectively). Age, B27, and psoriasis were the clinical variables associated with diagnosis of axSpA at 2 years in univariate analysis (all p<0.0001) and were included together with gender in multivariate analyses. Active and structural lesions typical of axSpA and SPARCC BME score ≥2 were each independently associated with diagnosis of axSpA at 2 years (OR(95% CI)- 6.8 (1.4–34.1) (p=0.02); 17.9 (2.2–146.6) (p=0.007); 4.9 (1.3–18.4) (p=0.02). With all variables simultaneously added to the model, only structural lesions were significantly associated.

Conclusions Assessment of both active and structural lesions on MRI may help determine which patients have axSpA with higher diagnostic certainty over time.

Disclosure of Interest None declared